Model: GELU Model: 4 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 612 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.8017. Min Act: -0.1700

Data Index: 1262050 (C4 (Web Text))

Max Activating Token Index: 769

Text #1

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.7078. Min Act: -0.1700

Data Index: 595699 (C4 (Web Text))

Max Activating Token Index: 305

Text #2

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.7391. Min Act: -0.1700

Data Index: 1199161 (C4 (Web Text))

Max Activating Token Index: 723

Text #3

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.6139. Min Act: -0.1700

Data Index: 335046 (C4 (Web Text))

Max Activating Token Index: 492

Text #4

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.6452. Min Act: -0.1700

Data Index: 447235 (C4 (Web Text))

Max Activating Token Index: 732

Text #5

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.5826. Min Act: -0.1700

Data Index: 181041 (C4 (Web Text))

Max Activating Token Index: 421

Text #6

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.6452. Min Act: -0.1700

Data Index: 89618 (C4 (Web Text))

Max Activating Token Index: 300

Text #7

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.6139. Min Act: -0.1700

Data Index: 295593 (C4 (Web Text))

Max Activating Token Index: 645

Text #8

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.6139. Min Act: -0.1700

Data Index: 394608 (C4 (Web Text))

Max Activating Token Index: 683

Text #9

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.5826. Min Act: -0.1700

Data Index: 1037435 (C4 (Web Text))

Max Activating Token Index: 755

Text #10

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.4886. Min Act: -0.1700

Data Index: 1554838 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.5200. Min Act: -0.1700

Data Index: 893694 (C4 (Web Text))

Max Activating Token Index: 843

Text #12

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.4886. Min Act: -0.1700

Data Index: 910621 (C4 (Web Text))

Max Activating Token Index: 153

Text #13

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.4259. Min Act: -0.1700

Data Index: 485460 (C4 (Web Text))

Max Activating Token Index: 322

Text #14

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.4573. Min Act: -0.1700

Data Index: 1135019 (C4 (Web Text))

Max Activating Token Index: 418

Text #15

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.4573. Min Act: -0.1700

Data Index: 1637935 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.4259. Min Act: -0.1700

Data Index: 1283754 (C4 (Web Text))

Max Activating Token Index: 968

Text #17

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.3946. Min Act: -0.1700

Data Index: 192748 (C4 (Web Text))

Max Activating Token Index: 976

Text #18

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.3632. Min Act: -0.1700

Data Index: 1227043 (C4 (Web Text))

Max Activating Token Index: 981

Text #19

Max Range: 3.8017. Min Range: -3.8017

Max Act: 3.3946. Min Act: -0.1700

Data Index: 1583922 (Python Code)

Max Activating Token Index: 53